Proposal of the S-score for measuring the performance of researchers, institutions, and journals in Indonesia - Science Editing
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
pISSN 2288-8063 Sci Ed 2018;5(2):135-141 eISSN 2288-7474 https://doi.org/10.6087/kcse.138 Case Study Proposal of the S-score for measuring the performance of researchers, institutions, and journals in Indonesia Lukman Lukman1, Muhammad Dimyati2, Yan Rianto1, Imam Much Ibnu Subroto3, Tole Sutikno4, Deden Sumirat Hidayat5, Irene M Nadhiroh6, Deris Stiawan7, Sam Farisa Chaerul Haviana3, Ahmad Heryanto7, Herman Yuliansyah6 1 Indonesian Institute of Science, Cibinong; 2Ministry of Research, Technology, and Higher Education, Jakarta; 3Department of Informatics Engineering, Universitas Sultan Agung, Semarang; 4Department of Electronic Engineering, Universitas Ahmad Dahlan, Yogyakarta; 5Research Center for Biology, Indonesian Institute of Science, Cibinong; 6Research Center for Science and Technology Development Studies, Indonesian Institute of Science, Jakarta; 7Department of Informatics Engineering, Universitas Sriwijaya, Indonesia, Palembang, Indonesia Abstract This study aimed to propose a tool for measuring the research performance of researchers, in- stitutions, and journals in Indonesia based on bibliometrics. Specifically, the output of this mea- Received: March 16, 2018 surement tool, referred to as the S-score, is described, as well as its implementation on the main Accepted: May 21, 2018 database portal in Indonesia. The S-score was developed by a focus group discussion. The fol- Correspondence to Lukman Lukman lowing 8 evaluation items for journal accreditation were analyzed in the development process: lukmanpdii@gmail.com journal title, aims and scope; publisher; editorial and journal management; quality of articles; writing style; format of PDF and e-journal; regularity; and dissemination. The elements of the ORCID Lukman Lukman S-score are as follows: number of journal article documents in Scopus, number of non-journal- https://orcid.org/0000-0001-9633-6964 article in Scopus, number of citations in Scopus, number of citations in Google Scholar, the h- Muhammad Dimyati index in Scopus, and the h-index in Google Scholar. The S-score yields results ranging from S1 https://orcid.org/0000-0002-2403-2619 Yan Rianto to S6. The above metrics were implemented on the Science and Technology Index, a database https://orcid.org/0000-0001-6056-6741 portal in Indonesia. The measurement tool developed through the focus group discussion was Imam Much Ibnu Subroto https://orcid.org/0000-0001-8311-8116 successfully implemented on the database portal. Its validity and reliability should be monitored Tole Sutikno consistently through regular assessments of S-scores. The S-score may be a good example of a https://orcid.org/0000-0002-1595-2915 Deden Sumirat Hidayat metric for measuring the performance of researchers, institutions, and journals in countries https://orcid.org/0000-0002-1847-8665 where most journals are not indexed by Scopus. Irene M Nadhiroh https://orcid.org/0000-0002-1417-6117 Deris Stiawan Keywords https://orcid.org/0000-0002-9302-1868 Academies and institutes; Accreditation; Bibliometrics; Focus groups; Indonesia Sam Farisa Chaerul Haviana https://orcid.org/0000-0003-1803-9803 Ahmad Heryanto https://orcid.org/0000-0003-2856-8135 Herman Yuliansyah https:/orcid.org/0000-0001-5401-8677 This is an open access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/ licenses/by-nc/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. http://www.escienceediting.org Copyright © 2018 Korean Council of Science Editors 135
Lukman Lukman et al. Introduction through the Indonesian national accredited journal database (http://arjuna.ristekdikti.go.id/). The performance of research- Universities in Indonesia are growing rapidly. As of February 6, ers, institutions, and journals was presented in SINTA. 2017, there were 4,498 colleges with 268,322 lecturers (Table 1) [1]. The Ministry of Research, Technology and Higher Educa- Indicators and evaluation items tion, as well as most universities and research institutions in Table 2 presents the following evaluation items, contents, and Indonesia, used the Scopus and Web of Science databases to scores used for journal accreditation by the Indonesian gov- evaluate the research performance of Indonesian researchers as ernment: journal title, aims and scope; publisher; editorial of 2017 [2]. Indonesia has not developed its own measurement and journal management; quality of articles; writing style; for- tool for research performance that can be used by policy-mak- mat of PDF and e-journal; regularity; and dissemination [4]. ers to evaluate researchers, institutions, researchers, and jour- An explanation of constants and how they are weighted to nals in various research fields [3]; therefore, a new measure- calculate the S-score is presented in Table 3: A, number of ment tool for assessing research performance in Indonesia is journal article documents in Scopus; B, number of non-jour- required. This study aims to propose a new metric, referred to nal-article documents in Scopus; C, number of citations in as the S-score, to measure the research performance of re- Scopus; D, number of citations in Google Scholar; and N, cur- searchers, institutions, and journals in Indonesia. This study rent divisor. The measurement items used to evaluate the per- proposes specific models, indicators, and metrics, and provides formance of researchers, institutes, and journals are presented the results of the implementation of these metrics on a data- in Fig. 1. base portal. The results will be useful for countries where many journals are not indexed in international citation databases, S-score and metrics for performance measurement such as Scopus or Web of Science. The S-score, which incorporated metrics for authors and in- stitutions, was formulated by weighting factors, as shown in Measurement Tools Table 3. The indicators and items for measuring research per- formance were formulated in two scenarios, which were ex- This tool for measuring research performance was proposed pressed as formula 1 and 2. The first scenario of the S-score through a focus group discussion in 2017. The expert group formula for researchers, without the h-index, was as follows: identified indicators based on the research products of re- (A × 40)+(B × 15)+(C × 4)+(D × 1) searchers and their citation frequency in Scopus and Google (1) N Scholar. The indicators and evaluation items were proposed for measuring the performance of not only authors but also of The second scenario of the S-score formula for researchers, institutions. Journals’ performance was measured by taking with the h-index, was as follows:V data from Indonesian journals indexed in Scopus, the Indo- (A × 40)+(B × 15)+(C × 1)+(D × 4)+(E × 16)+(F × 4) nesia national journal accreditation system, and citation fre- (2) N quency in Google Scholar. Based on these evaluation items, the S-score was proposed. The S-score formula for institutions was as follows: Data on research products from all lecturers and research- (A × 5)+(B × 2)+(C × 2)+(D × 1) ers in Indonesia were collected and entered into the Science (3) 10 and Technology Index (SINTA, http://sinta2.ristekdikti.go.id/ author) portal. The data collected from journals were matched If we use the maximum SINTA score of a researcher on Table 1. Number of universities, students, and lecturers in Indonesia in 2017 Universities Students Lecturer Public Private Total Public Private Total Public Private Total Universities 122 3,132 3,252 1,573,188 2,818,535 4,391,723 71,758 164,117 235,875 Religious universities 77 990 1,057 305,289 150,606 455,895 12,100 10,287 22,827 Service universities 177 0 182 107,028 0 107,028 9,619 0 9,619 Total 381 4,122 4,498 1,985,505 2,969,141 4,954,646 93,477 174,845 268,322 Data from Pangkalan Data Perguruan Tinggi. Kemenristekdikti [Internet]. Jakarta: Ristekdikti; 2017 [1]. 136 | Sci Ed 2018;5(2):135-141 http://www.escienceediting.org
Proposal of the S-score for measuring the performance of researchers, institutions, and journals in Indonesia Table 2. Evaluation items for journal accreditation by the Indonesia government Maximum Evaluation item Content score Journal title, aims, and scope Journal title is meaningful, precise, and short so that it is easily referenced. 3 Aims and scope of journal should be lucid and unique. The research field should be indicated. Publisher The publishing institute (professional organizations, universities, research and development institutes, and/or 4 institutes authorized for it) have the status of a legal entity, thus able to guarantee the continuity of funds and legal protection Editorial and journal management Reviewer 5 Management of quality of articles 2 Editorial board 3 Author guidelines 2 Quality of editing and formatting 2 E-journal management system (e-submission system) 3 Quality of article It fits the scope of the journal 4 Regional boundaries (international, regional, national, local) 6 Scientific originality of works 6 Contribution to the advancement of science 3 Citation 5 Primary reference source (journal, proceedings) ratio to other resources 4 Completeness of references 5 Analysis method 3 Conclusion 3 Writing style Representative article titles (straightforward and informative) 12 Inclusion of authors and affiliations (complete and consistent) 1 Abstract 2 Keyword 1 Structured description 1 Utilization of supporting documents (tables, figures, or supplements) 1 Reference citation style 1 Reference management (applications like Mendeley, etc.) 2 Terminology and language 2 Format of PDF and e-journal Format of PDF 8 Layout 1 Typography 1 PDF document resolution 2 Number of pages per volume 2 Journal website design 1 Regularity Regular publication 2 System of publishing order (volume, issue) 2 Page numbering 1 Retrieval in journal website (article, author) 1 Dissemination Count of unique visitors 4 Indexed in international databases (Scopus, Web of Science, DOAJ, etc.) 5 Unique identifier of articles (DOI) 2 Total 100 DOAJ, Directory of Open Access Journals. January 31, 2017 as the divisor (N), its value would be 102. results of the performance measurement will dynamically de- The corresponding graph is shown in Fig. 2. The results of the pend on the author registration process and data verification. measurements of author performance for all authors and in- The S-score for measuring the performance of journals is stitutions can be accessed at http://sinta2.ristekdikti.go.id. The classified into a range of S1–S6, as measured by the values of http://www.escienceediting.org Sci Ed 2018;5(2):135-141 | 137
Lukman Lukman et al. Table 3. Explanation of constants used to calculate the S-score suggested for measuring the performance of authors and institutes in Indonesia Assessment Constant explanation Additional information weight A No. of journal article documents in Scopus 40 B No. of non-journal-article documents in Scopus 15 C No. of citation in Scopus 4 If the no. of citations exceeds 1,000, then 1,000 will be used D No. of citations in Google Scholar 1 If the no. of citations exceeds 1,000, then 1,000 will be used E h-index in Scopus 16 F h-index in Google Scholar 4 N The divisor is a numerical result of a statistical calculation that takes into The current divisor value used is 102 (January 31, 2017) ccount the maximum SINTA score of a lecturer or researcher SINTA, Science and Technology Index. Measurement research performance Table 4. Accreditation classification of journals published in Indonesia adopt- ed by the portal Science and Technology Index Classification Accredited score Researcher Institution Indicators Journal performance performance performance S1 > 85 or indexed in Scopus S2 71 – 85 1. No. of publication in Scopus, 1. Journal title, aims and scope S3 61 – 70 Web of Science 2. Publisher 2. No. of non-journal articles 3. Editorial and journal management S4 51 – 60 documents in Scopus 4. Quality of articles Criteria S5 41 – 50 3. No. of citation in Scopus 5. Writing style 4. No. of citation in Google Scholar 6. Format of PDF and e-journal S6 31 – 40 5. No. of lecture and researcher 7. Regularity 8. Dissemination Fig. 1. Indicators and their evaluation items in Indonesia. 12,000 10,000 8,000 6,000 4,000 2,000 0 200 400 600 800 1,000 1,200 Fig. 2. Maximum Science and Technology Index score of researchers on Janu- ary 31, 2017. the 8 criteria used for the accreditation of Indonesian jour- nals, as presented in Table 2. The results of the classification by scores (S1–S6) can be seen in Table 4. Integration of metadata into the model The emergence of electronic journals and open access publish- ing has improved accessing digital data for each article, so that Fig. 3. Model of data integration in Indonesia. 138 | Sci Ed 2018;5(2):135-141 http://www.escienceediting.org
Proposal of the S-score for measuring the performance of researchers, institutions, and journals in Indonesia Fig. 4. Model of performance mapping model in Indonesia. Fig. 5. Mechanism of author verification and data collection in Indonesia. http://www.escienceediting.org Sci Ed 2018;5(2):135-141 | 139
Lukman Lukman et al. such data can be measured and made interoperable with vari- listed in Scopus or Web of Science. Therefore, a new local sys- ous databases and indexers such as Scopus, Web of Science, tem incorporating locally appropriate metrics is required. This Google Scholar, and others. Currently, thousands of Indone- is a good example of such a system, and it was possible to incor- sian journals have been published in electronic form, and porate the S-score in the Indonesian portal system (SINTA). most journals are open access [5]. This study proposed a new Therefore, these new metrics could enrich the Indonesian data- model for integrating existing data in the Scopus and Google base portal. Scholar databases to measure research performance in Indo- There were some limitations to this study. It was difficult to nesia through the Scopus API (application programming in- identify and match researchers through Google Scholar because terface) data retrieval mechanism, which can be incorporated some researchers did not have photos and the author’s name in into SINTA as shown in Fig. 3. A model of mapping and visu- the profile was not always the same as that in articles. Affilia- alizing research performance based on data already entered tions in Google Scholar and Scopus were often different from into SINTA databases can be seen in Fig. 4. Mapping perfor- the current affiliations. Therefore findig out a precise number of mance evaluation results in such a way can show profiles and publication from an institution is difficult and the data integrity ratings of researchers, institutions, and journals. should be pursued continuously. Another problem in the imple- mentation of the S-score in SINTA is normalization; it is chal- Implementation and data verification lenging to make comparisons among authors, disciplines, or in- The resulting indicators, formulas, and models were then test- stitutions because research products vary according to the re- ed against registered lecturers with a national lecturer number search category. Key items for evaluating performance were de- (NIDN) at universities and researchers listed as functional in- veloped in this study and implemented in the SINTA portal vestigators at research institutes. Lecturers and researchers based on a variety of items, including the number of citations in simply fill out the registration form at SINTA (http://sinta2. Google Scholar (Table 3); however, Google Scholar cannot be ristekdikti.go.id/author/) by filling in the lecturer’s identifica- used as a key item because the quality of data was not consistent. tion number (NIDN/unique number of employee), name, title, Scopus identifier and Google Scholar identifier. Lecturers and Conclusion researchers who sign up can correct any discrepancies in pub- lication data and the impact after synchronization, metadata In the future, SINTA needs to adopt a normalized measurement in Google Scholar and Scopus, as shown in Fig. 5. As of June method [6]. Normalization takes into account the following pa- 30, 2017, registered authors have verified as many as 32,218 rameters: the average number of citations per publication (ex- publications based on 25,472 data points harvested from 1,424 cluding self-citations); the percentage of publications without a institutions. The number of journals evaluated was 995 of the citation, the average number of journal citations, and the perfor- 1,807 journals registered through the Arjuna portal (http://ar- mance of research units in related fields around the world. Mea- juna2.ristekdikti.go.id/). Information on the implementation surements of research performance to be implemented within of journal performance metrics in SINTA is available from: the SINTA must consider interdisciplinary and transdisciplinary http://sinta2.ristekdikti.go.id/journals. Journal profiles have areas, as some measures may not be adequate for such areas. been created on that site based on citation performance, with The h-index and some of its modifications are useful for h-index values taken from Google Scholar; this information is quantifying the performance of researchers, similarly to the available at http://sinta2.ristekdikti.go.id/journals/detail?id= 664. rankings of universities, and this is necessary in a global envi- ronment [7,8]. Rankings will be announced periodically (4 Meaning of New Metrics and Its Limitations times a year) because real-time rankings may confuse policy- makers if rankings serve as a tool for rewards and penalties This study proposed a new metric, referred to as the S-score, to based on research performance. measure the performance of researchers, institutions, and jour- The journal classification (S1–S6) should be considered as a nals based on a set of evaluation items. The S-score was success- determinant of Indonesian journals. This ranking system may fully implemented in SINTA as a way to present performance be adopted by other local countries. Table 2 contains broadly results. Most evaluation items were designed through a focus applicable items for journal evaluation. The S-score of research- group discussion. The opinions of an expert group are an im- ers and institutions is also based on the number of publications, portant component of the development of new metrics or a citations, and h-index; therefore, this system will serve as a novel evaluation system in a country because the evaluation model to be referred to by other countries. system should be based on that country’s academic circum- A new metric, referred to as the S-score, for measuring the stances. Most scholarly journals published in Indonesia are not performance of researchers, institutions, and journals in Indo- 140 | Sci Ed 2018;5(2):135-141 http://www.escienceediting.org
Proposal of the S-score for measuring the performance of researchers, institutions, and journals in Indonesia nesia was proposed and successfully integrated into the SINTA Available from: http://forlap.ristekdikti.go.id/ portal. Its validity and reliability should be monitored consis- 2. Lukman L, Rianto Y, Nadhiroh IM. Development of na- tently through regular assessments of S-scores. This may be a tional indicator performance evaluation indicators and re- good example of a metric suitable for measuring the perfor- search mapping systems with Scientometrics model. Ja- mance of researchers, institutions, and journals in a country karta: LIPI; 2017. where most journals are not indexed in Scopus. 3. Lukman L, Yaniasih Y, Maryati I, Silalahi MA, Sihombing A. The strength of 50 Indonesian institutions: Scopus in- Conflict of Interest dexed publication profile. Jakarta: Ministry of Research, Technology and Higher Education; 2016. No potential conflict of interest relevant to this article was re- 4. Dikti D. Accreditation guidelines for scientific journal. Ja- ported. karta: Directorate General of Higher Education; 2014. 5. Lukman L. Index introduction: Scopus, Thomson Reuters, Acknowledgments DOAJ, Sinta for librarians, lecturers and researchers. Ja- karta: College Library Association; 2017. This study was supported by a research grant provided by the 6. Waltman L, Calero-Medina C, Kosten J, et al. The Leiden Insentif Sistem Inovasi Nasional (Insinas) Programme of the ranking 2011/2012: data collection, indicators, and interpre- Ministry of Research, Technology, and Higher Education of tation. J Am Soc Inf Sci Technol 2012;63:2419-32. https:// Indonesia Government (grant no. 62/P/RPL-LIPI/INSINAS-2/ doi.org/10.1002/asi.22708 VI/2017). 7. Hirsch JE, Buela-Casal G. The meaning of the h-index. Int J Clin Health Psychol 2014;14:161-4. References 8. Meyers MA, Quan H. The use of the h-index to evaluate and rank academic departments. J Mater Res Technol 2017; 1. Pangkalan Data Perguruan Tinggi. Kemenristekdikti [In- 6:304-11. https://doi.org/10.1016/j.jmrt.2017.09.004 ternet]. Jakarta: Ristekdikti; 2017 [cited 2017 Apr 8]. http://www.escienceediting.org Sci Ed 2018;5(2):135-141 | 141
You can also read